Electroencephalograms are medical tests used by doctors to determine any irregular brain activity in patients. The drawn-out process of manually classifying each of these signals can be time-consuming, costly, and even subjective among neurologists. We make use of the HMS dataset to classify spectrograms of brain activity into one of six abnormal brain activities. By applying basic data preprocessing techniques, we show the potential and power of using signal processing techniques like the Short-Time Fourier Transform and the Continuous Wavelet Transform in creating an accurate prediction model. By training on images of spectrograms and EEG signal plots, we create a three-model ensemble that can accurately predict brain abnormalities, easing the load off physicians. Our research showcases the potential of transforms in signal processing tasks